Xiaochun Cao

Compactness and Consistency: A Conjoint Framework for Deep Graph Clustering

OptMerge: Unifying Multimodal LLM Capabilities and Modalities via Model Merging

TrojanTO: Action-Level Backdoor Attacks Against Trajectory Optimization Models

MergOPT: A Merge-Aware Optimizer for Robust Model Merging

Qualitative and Quantitative Quality Assessment of Low-Light Enhanced Images: A Dataset and Benchmark Metric

Inverse Reinforcement Learning with Dynamic Reward Scaling for LLM Alignment

BRep: Graph-structured Brain Representation Learning via Parametric High-order Dependence Measures

Heterogeneous Federated PEFTs via Sparse Mixture-of-Experts

MoMCE: Mixture of Modality and Cue Experts for Multimodal Deception Detection

Explaining Multimodal LLMs via Intra-Modal Token Interactions

Reasoning as an Attack Surface: Adaptive Evolutionary CoT Jailbreaks for LLMs

Subset Selection-based Attribution Regularization for Rational and Stable Interpretability

Stability and Generalization of Split Learning : Sequential or Federated

PersGuard: Preventing Malicious Personalization in Text-to-Image Diffusion via Model Backdoors

Faster-VPS: Accelerating Object-Level Interpretation of Multimodal Foundation Models

RobustIT: Adapter-Centric and Attack-Agnostic Anti-Backdoor Instruction Tuning

RoboView-Bias: Benchmarking Visual Bias in Embodied Agents for Robotic Manipulation

Dynamic $k$-shot In-Context Learning

Simplify In-Context Learning

Where MLLMs Attend and What They Rely On: Explaining Autoregressive Token Generation

Disentangled Graph Spectral Domain Adaptation

Rethinking Joint Maximum Mean Discrepancy for Visual Domain Adaptation

A Closer Look at Graph Transformers: Cross-Aggregation and Beyond

Unveiling the Power of Multiple Gossip Steps: A Stability-Based Generalization Analysis in Decentralized Training

Do We Really Need Message Passing in Brain Network Modeling?

Lie Detector: Unified Backdoor Detection via Cross-Examination Framework

Ada-R1: Hybrid-CoT via Bi-Level Adaptive Reasoning Optimization

Focal-SAM: Focal Sharpness-Aware Minimization for Long-Tailed Classification

Decoupled Graph Energy-based Model for Node Out-of-Distribution Detection on Heterophilic Graphs

DUALFormer: Dual Graph Transformer

Towards Irreversible Attack: Fooling Scene Text Recognition via Multi-Population Coevolution Search

CamEdit: Continuous Camera Parameter Control for Photorealistic Image Editing

RoMa: A Robust Model Watermarking Scheme for Protecting IP in Diffusion Models

Improved Techniques for Optimization-Based Jailbreaking on Large Language Models

Ex-VAD: Explainable Fine-grained Video Anomaly Detection Based on Visual-Language Models

Targeted Low-rank Refinement: Enhancing Sparse Language Models with Precision

OT-Attack: Enhancing Adversarial Transferability of Vision-Language Models via Optimal Transport Optimization

Continual Model Merging without Data: Dual Projections for Balancing Stability and Plasticity

Vad-R1: Towards Video Anomaly Reasoning via Perception-to-Cognition Chain-of-Thought

Understanding the Stability-based Generalization of Personalized Federated Learning

Style-Coherent Multi-Modality Image Fusion

One Image is Worth a Thousand Words: A Usability Preservable Text-Image Collaborative Erasing Framework

Modeling Multi-Task Model Merging as Adaptive Projective Gradient Descent

LightSAM: Parameter-Agnostic Sharpness-Aware Minimization

Targeted Low-rank Refinement: Enhancing Sparse Neural Networks with Precision

CleanerCLIP: Fine-grained Counterfactual Semantic Augmentation for Backdoor Defense in Contrastive Learning

Few-shot Text Adversarial Attack for Black-box Multi-task Learning

No Access, No Safety: Free Lunch Adversarial Attacks on Black-box NLP Models